Chapter 10L’Aquila Earthquake: A Wake-Up Callfor European Research and Codes

Iunio Iervolino, Gaetano Manfredi, Maria Polese, Andrea Prota,and Gerardo M. Verderame

Abstract From the L’Aquila 2009 earthquake three issues, among others, stronglyemerged to be addressed for the engineered structures, at least in Europe. Theyare related to near-source effects, non-structural damage, and reparability. Althoughthey are well known since quite long time, still regulations seem giving little, ifany, practice-ready tools to account for them. In the chapter, evidences from theevent and scientific needs are briefly reviewed and discussed. The modest aim of thepaper is to stimulate debate and research in the light of next generation of seismiccodes.

Keywords Near-source • Directivity • Pulse-like records • Seismic hazard •Inelastic displacement ratio • Response spectrum • Non-structural elements •Infills • Reinforced concrete • Seismic assessment • Seismic design • Capacity •Codes • Damage • Collapse • Reparability • Substandard structures • Shearfailure • Soft-storey • Residual drift • Non-linear structural analysis

10.1 Introduction

The April 6, 2009 L’Aquila earthquake (MW 6.2) caused about three hundredsof fatalities, more than a thousand injuries, and extensive and severe damage tobuildings and other structures. About 66,000 residents were temporarily evacuated,and more than 25,000 were medium-term homeless.

The area of interest is known to be seismically active since a long time.Several events comparable in magnitude to this last earthquake, are reported by the

I. Iervolino (�) • G. Manfredi • M. Polese • A. Prota • G.M. VerderameDipartimento di Ingegneria Strutturale, Università degli Studi di Napoli Federico II,Via Claudio 21, 80125 Naples, Italye-mail: iunio.iervolino@unina.it; gaetano.manfredi@unina.it; mapolese@unina.it;andrea.prota@unina.it; verderam@unina.it

M. Fischinger (ed.), Performance-Based Seismic Engineering: Vision for an EarthquakeResilient Society, Geotechnical, Geological and Earthquake Engineering 32,DOI 10.1007/978-94-017-8875-5__10, © Springer ScienceCBusiness Media Dordrecht 2014

129

130 I. Iervolino et al.

national seismic catalogue. The main documented events, considering an estimatedmagnitude larger than 6.5, date to 1315, 1349, 1461, 1703 and 1915. In fact, the firstmodern-era seismic classification of L’Aquila refers to 1915, after the catastrophicAvezzano earthquake. The subsequent estimations of seismic hazard lead to thecurrent value of expected peak ground acceleration, or PGA, (for a return periodof 475 years) equal to about 0.26 g on rock. In fact, it is one of the largest seismichazard sites according to the national hazard map (http://esse1.mi.ingv.it/).

Analysis of buildings stock in the city of L’Aquila shows a percentage ofreinforced concrete buildings equal to 24 %, while the masonry buildings are 68 %;a percentage of 8 % refers to buildings of different typology. For what concernsthe age of construction, 55 % of buildings was built after 1945. Therefore, themost of existing buildings was built using some seismic provisions; neverthelesswith non-modern seismic standards. Moreover, seismic demand during the 2009earthquake was, locally, much larger than the design one. This seems also dueto near-source directivity effects. Generally, the seismic performance of buildingsstock in L’Aquila was considered unsatisfactory. In fact, in the 6 months after themainshock, the national department of civil protection organized a global surveyof all the buildings in the area affected by earthquake. About 80,000 field surveyswere performed by specialized teams. The results show that about 20,000 buildingsuffered of large structural damage, while about 10,000 buildings suffered of lightstructural damage and/or non-structural damage. Lacks and deficiencies of seismicdesign are believed to be responsible for such a bad performance, which is going tohave a large reconstruction cost for the country.

Starting from L’Aquila experiences, some remarks on possible improvementsof next generation of codes are discussed in the following. Of the many facetsof seismic risk, which a number of researchers studied after the earthquake, byfar the best documented event in Italy, three are those briefly discussed in thischapter: (i) directivity-related near-source effects of engineering interest and theirpredictability; (ii) seismic behaviour and structural dynamics’ influence of infillsin reinforced concrete structures; (iii) reparability and the possibility to explicitlyinclude this limit-state in design.

The reader may argue these are well known earthquake engineering topics sincequite some time, yet European codes, at least, are somewhat lacking in their respect,while level of scientific knowledge is such they may be considered in the nextgeneration of seismic regulations.

10.2 Near-Source Pulse-Like Engineering Issues

Near-source (NS), or directivity, effects in ground motion depend on the relativeposition of the site with respect to the fault rupture, and typically appear by means oflarge velocity pulses concentrating energy in the starting phase of the fault-normal,or FN (i.e., normal to the rupture’s strike), component. This results in waveformsdifferent from ordinary ground motion recorded in the far field, or in geometrical

10 L’Aquila Earthquake: A Wake-Up Call for European Research and Codes 131

0.10.1

0.10.1

0.20.2

a

b

42.6

CSO

AVZCLN SUL

CDS

CHTFMG

0.10.1 0.30.3 0.20.2

0.20.2

0.50.50.40.4

0.40.40.30.3

0.30.30.20.20.10.1

0.10.1

0.10.1

ANTMTR

AQAAQAAQGAQG

ORCORC

AQVAQVAQUAQU

GSAGSAGSGGSG

AQKAQK

SBC

42.4

42.2

42

Lat

itud

e [°

]

Longitude [°]

AQU Fault Normal

Time [sec]

Vel

ocity

[cm

/s]

5

500

-50

500

-50

500

-5010 15 20

41.8

13 13.2 13.4 13.6 13.8

EpicenterPulse ProbabilityClstd = 30km

14 14.2

Fig. 10.1 Near-source stations in L’Aquila 2009 earthquake and pulse occurrence probabilitycontours (according to the model of Iervolino and Cornell 2008) together with actual identified(in bold) velocity pulses (a); impulsive signal of AQU station in the same earthquake; from top tobottom: velocity time history, extracted pulse and residual velocity (b)

conditions not favorable with respect to directivity. Near-source pulse-like traceswere found in the records of L’Aquila earthquake (Chioccarelli and Iervolino 2010).

In Fig. 10.1a the seismic stations, which have recorded the event close to thesource, are shown and superimposed to a probability model for occurrence of pulse-like records (Iervolino and Cornell 2008). In bold there are the stations wherethe records, in which pulses supposed to be originated by directivity, were found;consistency may be observed. In Fig. 10.1b, for one accelerometric station (AQU),the original velocity signal is shown together with the extracted pulse and theresidual ground motion once the pulse is removed (according to the algorithm inBaker 2007).

In Chioccarelli and Iervolino (2010), analyzing the NGA database (http://peer.berkeley.edu/nga/), it was found that the three main characteristics of pulse-likerecords of earthquake engineering interest are:

132 I. Iervolino et al.

10−2 10−1

4.5

3.5

2.5

1.5

0.5

4

3

2

1

0

CR

Pulse - FNPulse - FPNon Pulse - FN

R = 4

101

100

10−1

Sa/P

GA

10−2

10−3

Average Non PulseAverage Pulse (1s<Tp<2s

T (s) T/Tp

101 0 1 20.5 1.5100

a b

Fig. 10.2 Elastic 5 % damped spectra for FN pulse-like with 1 s < TP < 2 s and ordinary records(a); empirical CR for FN pulse-like records, for their fault-parallel (FP) components, and forordinary records, for a strength reduction factor R D 4 (b) (Adapted from Iervolino et al. 2012)

1. pulse-like signals are characterized by fault normal records generally strongerthan both fault parallel components and non-pulse-like ground motions;

2. fault-normal pulse-like records are characterized by a non-standard spectralshape with an increment of spectral ordinates in a range around the pulse period(Tp), Fig. 10.2a;

3. inelastic-to-elastic seismic spectral displacement ratio (CR) for pulse-like recordscan be 20–70 % higher than that of ordinary motions depending on the non-linearity level; such increments are concentrated in a range of period between 30and 50 % of pulse period of each record, Fig. 10.2b.

These points show that near-source directivity is of interest to earthquakeresistant design, which has to be adjusted for near-source because: (i) traditionalhazard assessment may be unable to predict (1) and (2), that is, the elastic peculiarfeatures of pulse-like records; (ii) point (3) shows that current static design, basedon equal displacement rule, may fail when pulse-like records are concerned; (iii) itis not implicit in current design practice that safety margins are consistent betweenordinary and pulse-like records.

Regarding (i): in near-source conditions the probabilistic seismic hazard analysis(NS-PSHA), expressed in terms of mean annual frequency of exceeding a spectralacceleration (Sa) level, or �Sa, is a linear combination of two terms, which accountfor the occurrence or the absence of the pulse, �Sa,NoPulse and �Sa,Pulse, weighted bythe pulse occurrence probability, Eq. (10.1). Moreover, NS-PSHA requires dealingwith two more tasks, which are not faced in traditional hazard analysis: pulse periodprediction; and pulse amplitude prediction.

�Sa.x/ D �Sa;NoP ulse.x/ C �Sa;P ulse.x/ (10.1)

10 L’Aquila Earthquake: A Wake-Up Call for European Research and Codes 133

100100

80

60

(Sa N

S-PSH

A−S

a PSH

A)/Sa

PSH

A [%

]

SaPSHA M=[5 6 7]

SaNS-PSHA M=[5 6 7]

M=[5 6 7]M=7M+=7

SaNS-PSHA M=7

SaPSHA M=7

SaNS-PSHA M=7

SaPSHA M=7 40

20

010−110−2 101100 0 2 4 6 8 10

T [sec] T [sec]

Sa [g]

10−2

10−4

+

+

a b

Fig. 10.3 475 years UHSs with modified and classical PSHA (a); increments due to directivityeffects (b) (Adapted from Chioccarelli and Iervolino 2013)

4.5

3.5

2.5

1.5

0.5

4

3

2

1

0

CR

Actual Data

Actual Data + σCR

Fitted Functional Form + σCR

Fitted Functional Form

R = 4

10−1

Sa/P

GA 10−2

10−3

UHSNS-PSHA

CIS

T [sec]T/Tp

010.10.01 1 20.5 1.5

a b

Fig. 10.4 (a) Near-source possible design scenarios (Chioccarelli and Iervolino 2013);(b) inelastic displacement ratio functional form for pulse-like records fitted to actual data (Iervolinoet al. 2013)

As an example, in Fig. 10.3a near-source uniform hazard spectra (UHS, 10 % in50 years), are compared to UHS from traditional PSHA for different cases of mag-nitude distribution and source-to-site-geometry for a strike-slip fault (Chioccarelliand Iervolino 2013). In Fig. 10.3b increments of NS-PSHA with respect to PSHAare given. It is to observe that these may be significant. It also appears the UHS isonly able to capture the NS amplitude, not the peculiar spectral shape of Fig. 10.2a.Therefore, it seems necessary to find alternate solutions to get design seismicactions.

One possible approach is that proposed in Chioccarelli and Iervolino (2013) andnamed close-impulsive spectrum (CIS), conceptually consistent with the conditionalmean spectrum Baker (2011). CIS is derived from disaggregation of NS-PSHAin terms of magnitude, distance from the source, and pulse period. In Fig. 10.4a,starting from the hazard for a 0.5 s Sa, the shape of CIS is reported for a case in

134 I. Iervolino et al.

which the design scenario is 0.7 s, 5, and 10 km, in terms of Tp, magnitude, andsource-to-site distance, respectively. The figure also shows the UHS for the sameexample, to appreciate differences and how CIS reproduces a ‘bump’ around thedesign pulse period.

Regarding point (ii), it was found in Iervolino et al. (2012) that the inelasticdisplacement ratio of near-source pulse like records, requires a specific functionalform to fit observed data as a function of T/Tp, which is substantially differentfrom ordinary records. In fact, the functional form in Eq. (10.2), which consistsof adding two opposite bumps in two different spectral regions to the traditionalhyperbolic format of CR in codes (e.g., FEMA 2005), may be suitable. In Fig 10.4bthis functional form is plotted (also plus one standard deviation) against data ofFig 10.1b once the ™ coefficients are fitted.

CR D 1 C �1 � �Tp=T�2 � .R � 1/ C �2� � �Tp=T

� � expn�3 � �ln �T=Tp � 0:08

��2o

C �4 � �Tp=T� � exp

n�5 � �ln �T=Tp C 0:5 C 0:02 � R

��2o(10.2)

10.3 Non-structural Elements and Their Influenceon Structural Assessment/Design

Infill walls are usually employed in reinforced concrete (RC) buildings for partitionuse and for thermal/acoustic insulation. Hence, they are considered as non-structuralelements; nevertheless, post-earthquake damage observation, experimental andnumerical research, showed that their influence on seismic behaviour of RCbuildings can be not negligible at all.

Modern seismic codes (e.g., FEMA 2000; CEN 2004; DM 2008) prescribe toaccount for the possible influence of infills on seismic behaviour of RC frames,both at local and global level. In particular, according to Eurocode 8 (CEN 2004), ifwalls take at least 50 % of the base shear from a linear analysis, the interaction ofthe structure with the masonry infills may be neglected. This may be taken to implythat it is allowed then to disregard the infills in the structural model. However, this isnot always a safe assumption. An asymmetric layout of the infills in plan may causetorsional response to the translational horizontal components of the seismic action;so, according to Eurocode 8 part 1, infills with strongly asymmetric or irregularlayout in plan should be included in a 3D structural model and a sensitivity analysisof the effect of the stiffness and position of the infills should carried out. Eventhough the awareness about this issue in earthquake engineering is not very recent,it is likely to state that practically no existing RC building was designed accountingfor the presence of these elements.

This latter observation becomes a key issue in countries, such as Italy, in whichmost of the building stock was realized before 1990s. It has to be noted thatnot even limit state approach was provided in the Italian regulations until 1996,

10 L’Aquila Earthquake: A Wake-Up Call for European Research and Codes 135

0%

20%

40%

60%

80%

100%

<1919 19-45 46-61 62-71 72-81 82-91 >1991

Fig. 10.5 2001 census ISTAT data for L’Aquila. The construction age intervals on the abscissaidentify the percentage of buildings to add to the preceding bar to get the bar they correspond to,that is, to obtain the cumulative distribution (Ricci et al. 2011a)

Fig. 10.6 (a) Shear failure of column adjacent to partial infilling panels; (b) diagonal crackingfailure in concrete joint panel (Ricci et al. 2011a)

see Ricci et al. (2011a) for details. Figure 10.5 shows the empirical cumulativedistribution of the age of construction of L’Aquila buildings that can be approx-imately considered representative of most of the Italian region in terms of urbandevelopment.

Observation of damage on RC structures after L’Aquila event emphasized theimportance of taking into account the contribution of non-structural elements suchas infills. Most of the observed damage was localized in infill panels, and theirstiffness and strength contribution in some cases preserved the RC buildings fromstructural damage. On the other hand, it is to note that infill irregular distributionin plan and elevation can be addressed as one of the main causes of the structuralcollapses observed during in-field campaigns, together with brittle failure mode ofcolumns and beam-column joints (see Fig. 10.6).

136 I. Iervolino et al.

The lack of capacity design and shear-flexure hierarchy can lead to brittle failuresin primary elements (De Luca and Verderame 2013), and interaction with masonryinfills can favor the occurrence of them. Effect of infills on the whole structuralbehavior depends on different factors that are briefly recalled in the following.

First, the evaluated fundamental period of the infilled structure (Ricci et al.2011b) changes significantly if compared to the period of the bare frame model,since the structure is significantly stiffer. Furthermore, infill presence can changeregularity characteristics of the structure in plan and in elevation and consequentlycan affect the mode of vibration of the building. Second, infills are characterizedby a brittle behavior and the high contribution in strength they provide to a RCbuilding suddenly decreases for low values of drift. On the other hand, it is worth tonote that infill impact on the structural performances of a building becomes a criticalissue when RC primary elements are designed according to obsolete criteria and thestructure is characterized by an insufficient global and local ductility. The latter canbe observed when seismic performances of contemporary and existing infilled framestructures are compared (Dolsek and Fajfar 2005). If infill distribution is irregular inplan or elevation, their contribution introduces a source of irregularity (e.g. Pilotiseffect) and the possibility to register a soft-storey mechanism is dramaticallyincreased, especially when no capacity design criterion has been employed in thedesign of the bare frame.

Mechanical properties of the infills represent another critical factor that can varythe effects on the performances of the whole structure, since they are considerednon-structural elements and their properties, not systematically checked, can varysignificantly because of the local building practice. Regarding this latter issue,it is important to stress the relative weight that infills have with respect to themechanical properties of the bare frame. Because of all the variables considered(seismic intensity level, old or contemporary design approach, distribution andmechanical properties of the infills) it is tough to say if structural contribution ofthese “non-structural” elements increases or decreases the overall seismic capacityof the building (Ricci et al. 2012).

In (Verderame et al. 2011) one of the few collapsed buildings after L’Aquilaearthquake is assumed as case study; the building collapsed as a result of asoft storey mechanism at the first level (see Fig. 10.7). Observed damage pointsto collapse as a result of a brittle failure mechanism. Given the likely scenariocollapse inferred by observed damage, an analytical model of the building was builttaking into account nonlinear behavior of the infills; local interaction with columnswas also considered by means of a three strut macro-model. Two parametrichypotheses based on Italian code prescriptions were assumed for infill mechanicalproperties. Time history analyses were carried out assuming as seismic input thethree components of the real registered signals during the mainshock in the vicinityof the case study structure. Given the brittle failure highlighted by damage, besidecapacity models suggested by codes, other shear failure mechanisms (sliding shearfailure) not typical for columns were considered.

Numerical results seem to confirm the collapse scenario inferred by damageobservation; the lack of proper structural and executive details was addressed as

10 L’Aquila Earthquake: A Wake-Up Call for European Research and Codes 137

Fig. 10.7 (a) Pre-event view of case study building, placed in Pettino (L’Aquila) (from VirtualEarth); (b). Collapsed building (Verderame et al. 2011)

the main cause that made a critical issue the local interaction between columns andinfills, other than the strong vertical component registered.

The case-study structure, even if characterized by structural peculiarities, repre-sents in itself a lesson for future code provisions and, above all, it emphasizes aneffect typically disregarded in conventional assessment procedures.

10.4 Is a Reparability Limit State Needed and Feasible?

One of the most controversial problems in the aftermath of damaging earthquakesis the lack of agreed and transparent policies for acceptable levels of safety, aswell as of advanced technical standards for repair and/or strengthening of damagedbuildings. In fact, the technical difficulties for the assessment of the safety loss ofdamaged buildings and for the choice of the most appropriate method for repairand/or strengthening of damaged elements, avoided the development of sound andagreed re-occupancy criteria meeting engineering consensus on the different aspectsto be considered.

The criteria for reconstruction funds assignments after L’Aquila earthquakedistinguish between buildings classified as B/C (generally having slight damages)by the damage survey forms used by the Italian Civil Protection (Baggio et al. 2007),with respect to those classified as E (heavy structural damage). However, if the usualtagging procedures, based on an expert assessment of damage level and extent bya team of experienced practitioners, are deemed acceptable in an emergency phasein order to establish building usability, they cannot be considered as the only toolfor choosing intervention categories and assigning grantable funds or insurancerefunds for damaged buildings. A proper approach should rely on a performancebased policy framework, as suggested in FEMA 308 (FEMA 1998b), wherebuilding’s performance index and performance loss due to earthquake damageare the parameters considered to drive decisions among the alternatives of simplyaccepting the building as it is (insignificant damages), repairing or upgrading it.

138 I. Iervolino et al.

Fig. 10.8 (a) Dense digital elevation model of the building in Pianola (AQ) obtained with laser-scanner 3D. (b) Examples of damages on the elements at the first storey (Polese et al. 2011)

Only partly in line with this approach, post L’Aquila regulations implicitlyassociate E buildings with medium-high performance loss, and the repairing andupgrading of buildings becomes mandatory when the performance index (ratio ofthe PGA corresponding to building collapse versus the design PGA in the nominalbuilding life), evaluated for the intact building, is lower than 60 %. Moreover, withthe purpose of avoiding length and costly investigations and analyses for heavilydamaged buildings, an evaluation policy was introduced considering the amount ofresidual drifts as discriminative parameter for reparability. In particular, the coderegulation OPCM 3881 (2010) establishes that it is possible to avoid demonstrationof the economic convenience of demolishing and re-building if there are permanentdrifts �1.5 % for at least 50 % of the columns at the same storey. Such value issomewhat greater than current literature values; for example, in ATC 58-1 (2011)for a residual drift of 1 % it is hypothesized that major structural realignment isrequired to restore safety margin for lateral stability, and the required realignmentand repair of the structure may not be economically and practically feasible (i.e., thestructure might be at total economic loss). However, considering the typical shearsliding mechanism at the beam-column interface of existing under-designed Italianbuildings, due to masonry infill interactions with columns (Verderame et al. 2011),residual drift is often related to a local damage of some (typically external) columns,and larger permanent drifts may be expected, as reported in Polese et al. (2011). Infact, maximum residual drifts measured for two buildings in L’Aquila (Fig. 10.8 andTable 10.1), that were severely damaged at the first storey, is well beyond 1.5 % inboth cases. However, the real measures for these two buildings show that it is verydifficult to comply with the requirements of OPCM 3881 (2010) (at least 50 % ofcolumns in a storey with residual drifts > 1.5 %) even for very severely damagedbuildings.

10 L’Aquila Earthquake: A Wake-Up Call for European Research and Codes 139

Table 10.1 Permanent drifts(™A and ™B on orthogonalsides of columns) evaluated atthe first storey for twodamaged buildings in Pettino(AQ) and in Pianola (AQ)

Pettino (AQ) Pianola (AQ)

Col. ID ™B (%) ™A (%) Col. ID ™B (%) ™A (%)

01 C0.74 – 01 C1.04 �0.5602 C0.13 – 05 C0.25 C0.6404 C0.96 – 13 �1.79 C0.6006 C0.51 – 19 C0.77 C0.3507 C2.52 C0.40 17 C1.89 �0.5432 �0.93 C0.19 14 C0.29 0.00

Only perimeter columns, where it was possible to measureresidual deformations, are indicated (Polese et al. 2011)

IN2REC

RECkIN2k

for Teq ≥ TcµcapCbRECSa ⋅=

for Teq < TcTc

TeqRECSa = Cb ⋅ (µcap−1) +1⋅

Sd

Sa

Fig. 10.9 Application of IN2method for determiningREsidual Capacity in theintact (REC) and damagedstate (RECk) (Polese et al.2013)

Obviously, there is a need to further investigate on the relationship betweenresidual drifts, damages and performance loss in a performance-based policyframework. If varied vulnerability is to be considered in a consistent quantitativeassessment framework, analytical modeling of building performance loss, account-ing for building damage and residual drifts, is preferable. Building performance inthe pre-event, damaged and eventually restored/upgraded state may be investigatedwith nonlinear static analysis considering proper variation of element’s force-deformation relationships, as outlined in FEMA (1998a).

The seismic behavior of damaged buildings, and the relative seismic safety,may be adequately represented by their seismic capacity modified due to damage,the so called REsidual Capacity (REC). In the framework of a mechanical basedassessment of seismic vulnerability, REC may be evaluated based on pushovercurves obtained for the structure in different (initial) damage state configurationsand accounting for possible residual drifts. In particular, residual capacity RECSa

is defined, for each global damage state Di, as the minimum spectral acceleration(at the period Teq of the equivalent single degree of freedom system) correspondingto building collapse, and can be determined, among other approaches, with the IN2method (Dolsek and Fajfar 2004), see Fig. 10.9.

140 I. Iervolino et al.

A preliminary application Polese et al. (2013) shows that the adoption ofnonlinear static analyses for damaged building, with explicit consideration of thedamage and residual drifts in the main resisting elements, allows a consistentassessment of the building safety factor and performance loss with respect to intactstructure. In this application, suitable modification factors for moment rotationplastic hinges of the columns have been calibrated based on a number of cyclic testsperformed by the authors on non-conforming columns. However, models to predictthe modification factors based on a wider number of experimental tests availablein the literature (on columns representative of existing RC buildings of EuropeanMediterranean regions) are currently under investigation (Di Ludovico et al. 2012).

Referring to the case study, an interesting conclusion is that while for structuresthat have been slightly or moderately damaged the ultimate deformation capacitydoes not significantly change with respect to the undamaged structure, at thesame time these structures are more deformable (having higher Teq) and havea lower ductility capacity, �cap, with a consequent lower residual capacity. Thiscircumstance is reflected in the increasing performance loss, that reaches valuesnearly of 10 % for slight damage, and up to 20 % for a structure that has reached amoderate damage state due to an hypothetical mainshock.

A key issue for a proper assessment of performance loss, to be further developed,is the experiment-based characterization of typical existing RC columns perfor-mance in the intact and damaged states, distinguishing their possible behaviormodes (e.g. flexural, flexure/shear, sliding shear etc.) and significant damage levels.Moreover, in order to verify the effect of alternative retrofit solutions, also theinfluence of local retrofit interventions on such elements should be evaluated.

10.5 Conclusion

A few research opportunities and codes’ needs were highlighted starting from theevidence of L’Aquila 2009 earthquake. In fact, analysis of records and damages tothe built environment shows that the earthquake engineering research appears closeto be capable of a practice-ready implementation on some issues, while other stillrequire investigations, experimental tests, and competency building.

Three topics were found especially significant and briefly reviewed. In particular,near-source pulse-like effects, infills of reinforced concrete structures, and repara-bility, were analysed with respect to design and assessment.

On the near-source seismic demand side, it appears that there is a need foraccounting for the peculiar pulse-like features in the elastic (i.e., seismic hazard)and inelastic range; state-of-the-art tools are available, while current provisions maybe unable to capture those.

The influence of infills on the structural behaviour of engineering structureshas to be considered in the seismic design for two main reasons: (1) infills candramatically change the seismic behaviour determining unpredictable collapsemodes with respect to the bare structure; (2) in a consequence-based framework,

10 L’Aquila Earthquake: A Wake-Up Call for European Research and Codes 141

a reduction of infills damage can increase the immediate occupancy conditions andreduce the economic losses due to non-structural damage.

Finally, the definition of a reparability limit state determines potential improve-ments in seismic response; e.g., the definition of a clear condition of repair/upgradeconvenience in the post-event situation, and new criteria in the choice of alternativeretrofit solutions in a multi-criteria decision making method approach.

Acknowledgements The studies presented in this paper were developed within the activities ofRete dei Laboratori Universitari di Ingegneria Sismica (ReLUIS) for the research program fundedby the Dipartimento della Protezione Civile (2010–2013).

References

ATC 58-1 75% Draft (2011) Seismic performance assessment of buildings volume 1 – methodol-ogy. Applied Technology Council, Redwood City, California

Baggio C, Bernardini A, Colozza R, Di Pasquale G, Dolce M, Goretti A, Martinelli A, Orsini G,Papa F, Zuccaro G, Pinto AV, Taucer F (2007) Field manual for post-earthquake damage andsafety assessment and short term countermeasures (AeDES), EUR 22868 EN – 2007. JointResearch Center, Ispra

Baker JW (2007) Quantitative classification of near-fault ground motions using wavelet analysis.Bull Seismol Soc Am 97(5):1486–1501

Baker JW (2011) Conditional mean spectrum: tool for ground motion selection. J Struct Eng137(3):322–331

Chioccarelli E, Iervolino I (2010) Near-source seismic demand and pulse-like records: a discussionfor L’Aquila earthquake. Earthq Eng Struct Dyn 39(9):1039–1062

Chioccarelli E, Iervolino I (2013) Near-source seismic hazard and design scenarios. Earthq EngStruct Dyn 42(4):603–622

Comité Européen de Normalisation (CEN) (2004) Eurocode 8 – design of structures for earthquakeresistance – Part 1: General rules, seismic actions and rules for buildings. EN 1998-1, CEN,Brussels

De Luca F, Verderame GM (2013) A practice-oriented approach for the assessment of brittlefailures in existing RC elements. Eng Struct 48:378–388

Decreto Ministeriale (DM) del 14 gennaio (2008) Approvazione delle nuove norme tecniche per lecostruzioni. G.U. n. 29 del 4/2/2008 (in Italian)

Di Ludovico M, Polese M, Gaetani D’Aragona M, Prota A (2012) Plastic hinges modificationfactors for damaged columns. In: 15th world conference on earthquake engineering – 15thWCEE, Lisbon, Portugal, 24–28 Sept 2012

Dolšek M, Fajfar P (2004) IN2- a simple alternative for IDA. In: 13th world conference onearthquake engineering. Vancouver, BC, Canada, 2004; Paper no. 3353

Dolšek M, Fajfar P (2005) Simplified non-linear seismic analysis of infilled reinforced concreteframes. Earthq Eng Struct Dyn 34:49–66

Federal Emergency Management Agency (1998a) FEMA 306: evaluation of earthquake damagedconcrete and masonry wall buildings – basic procedures manual. Federal Emergency Manage-ment Agency, Washington, DC

Federal Emergency Management Agency (1998b) FEMA 308: repair of earthquake damagedconcrete and masonry wall buildings. Federal Emergency Management Agency,Washington, DC

Federal Emergency Management Agency (2000) FEMA 356: NEHRP guidelines for the seismicrehabilitation of buildings. Federal Emergency Management Agency, Washington, DC

142 I. Iervolino et al.

Federal Emergency Management Agency (2005) Improvement of nonlinear static seismic analysisprocedures. Report FEMA 440. Washington DC, US

Iervolino I, Cornell CA (2008) Probability of occurrence of velocity pulses in near-source groundmotions. Bull Seismol Soc Am 98(5):2262–2277

Iervolino I, Chioccarelli E, Baltzopoulos G (2012) Inelastic displacement ratio of near-sourcepulse-like ground motions. Earthq Eng Struct Dyn 41(15):2351–2357

OPCM 3881 (2010) Ulteriori interventi urgenti diretti a fronteggiare gli eventi sismici verificatisinella regione Abruzzo il giorno 6 aprile 2009 (in Italian)

Polese M, Prota A, Manfredi G, Dolce M (2011) Spostamenti residui in edifici in c.a. danneggiatidal sisma. XIV Convegno Anidis, Bari, Italy (in Italian)

Polese M, Di Ludovico M, Prota A, Manfredi G (2013) Damage-dependent vulnerability curvesfor existing buildings. Earthq Eng Struct Dyn 42(6):853–870

Ricci P, De Luca F, Verderame GM (2011a) 6th April 2009 L’Aquila earthquake, Italy – reinforcedconcrete building performance. Bull Earthq Eng 9(1):285–305

Ricci P, Verderame GM, Manfredi G (2011b) Analytical investigation of elastic period of infilledRC buildings. Eng Struct 33(2):308–319

Ricci P, De Risi MT, Verderame GM, Manfredi G (2012) Influence of infill presence and designtypology on seismic capacity of RC buildings: sensitivity analysis. In: 15th world conferenceof earthquake engineering, Lisbon, Portugal

Verderame GM, De Luca F, Ricci P, Manfredi G (2011) Preliminary analysis of a soft-storeymechanism after the 2009 L’Aquila earthquake. Earthq Eng Struct Dyn 40(8):925–944